Project Summary Spinal muscular atrophy (SMA) is a common, frequently fatal, neuromuscular disorder caused by mutations in the Survival of Motor Neuron 1 (SMN1) gene and, consequently, a paucity of the SMN protein. In humans, an almost identical copy gene, SMN2, is unable to fully compensate for loss of SMN1 owing to a splicing defect and thus an inability to express sufficient protein to stave off disease. In the two decades that we have researched SMA much progress has been made, from the identification of the disease gene and the description of its protein to the generation of pre-clinical models and, most recently, the approval of Spinraza, a promising drug that raises SMN levels and thus thwarts the inevitable paralysis and frequent death associated with SMA. While Spinraza, in particular, raises considerable optimism for SMA patients, significant challenges remain and, in our minds, stem from two crucial deficiencies. First, despite the milestones achieved, how low SMN protein evolves into the SMA phenotype, selectively triggering motor neuron death and preferentially disabling the neuromuscular system remains a singular mystery. This is especially perplexing considering SMN's most widely-cited function of orchestrating the splicing cascade. Identifying mediators that provide a logical explanation for why splicing defects cause SMA or, uncovering additional, more disease-relevant SMN functions is therefore not only mechanistically but also therapeutically relevant. Second, while it is clear that administering Spinraza provides immediate benefit to patients, it is premature to make a determination of the long-term outcome of such treatment; the drug is selectively delivered to the CNS, raising questions about the effects of chronic low SMN in the periphery. Besides, the strategy of raising SMN appears inadequate in the symptomatic patient. Here we describe 3 related sets of experiments that address the deficiencies identified above. Aim 1 proposes to define disease-relevant mechanisms by exploiting a novel line of SMA mice in which early mortality, motor neuron loss and a severe phenotype are replaced by prolonged survival, intact motor neurons and a decidedly mild phenotype. We hypothesize that a spontaneous mutation in a chaperone protein that the mice express suppresses the SMA phenotype. We will confirm and extend this finding to determine how the chaperone modulates the effects of low SMN. In aim 2, we will examine the potential long- term adverse effects of persistently low levels of SMN in muscles of model mice expressing normal protein in the CNS. Such rodents represent a pre-clinical model of SMA patients administered Spinraza. We propose that chronic low SMN in skeletal muscle has a profoundly negative impact on the health of the tissue and contributes to the overall SMA phenotype. In aim 3, we will determine if the disease-causing effects of low SMN in muscle can nevertheless be mitigated upon restoring protein post-symptomatically. Reversing such defects will inform the manner in which current treatments may have to be modified to prove more potent. Our study thus addresses important mechanistic as well as clinical aspects of SMA.